When I watch videos of probes landing on Mars after cutting the parachutes off, they show the lander doing final orientation adjustments before touching down.

The engines don't seem to have a vectoring range, but I don't know if that's an omission of the animation or it reflects reality.

So I am wondering how does the probe correct its pitch, yaw and roll.

note: My question is about attitude control of recent Mars landers post-parachute, not about various animations that are out there. There have been many landing schemes on Mars, it's not necessary to go through every one, but if it's possible to focus on the more recent ones that would be most helpful.

As an example this answer links to Spaceflight 101 which says:

They are operated in pulse mode to arrest the craft’s horizontal and vertical velocity and also keep it in the proper orientation for landing, achieved through differential pulsing of the thrusters to actively control the pitch, yaw and roll of the lander.

but it's not at all clear how that actually works.

ESA and Roscosmos:



  • 2
    $\begingroup$ Welcome to Space Exploration SE! Can you add a link to the video? $\endgroup$
    – BlueCoder
    Nov 25, 2018 at 21:15
  • $\begingroup$ Welcome to Space! As there have been several different missions which have landed on Mars -- using different ways -- it would be helpful if you could identify which mission it is. Thanks! $\endgroup$
    – DrSheldon
    Nov 25, 2018 at 23:34
  • $\begingroup$ This is a good question! In this answer there is some information from Spaceflight 101 for InSight, but I don't think it is sufficient to answer your question. I've made some edits to make it clearer your question is not about various videos. $\endgroup$
    – uhoh
    Nov 26, 2018 at 0:15

1 Answer 1


There are 12 thrusters on InSight, mounted around the outer edge of the lander, all pointed in fixed generally downward orientation. The thrusters have electrically operated valves which can be operated in quite short pulses - the thruster can fire for a fraction of a second at a time.

By controlling the rate and/or duration of the pulses differently on different thrusters, the lander can produce more upward force on one side or the other, which causes it to tilt in the desired direction. This is "differential pulsing". I believe similar systems were used on Viking and MSL; not sure about Spirit & Opportunity.

During the terminal phase of descent, when the lander needs to be braking hard, the thrusters will be mostly on, with them turning off briefly on one side or another to control attitude by letting that side dip.

This gives pitch and yaw control. Roll control is a little less critical for descent, but it looks to me like the groups of thrusters at each corner of the lander are canted slightly apart from one another:

enter image description here

So differentially pulsing the thrusters in each group would give a little roll control.

The Spaceflight101 description suggests that the thrusters may be throttlable, but it's not clear to me if that means producing a time-averaged thrust level by pulsing, or a constant lower thrust by holding the propellant valve partly open; Aerojet's datasheet doesn't describe that exact model. Differential throttle attitude control works essentially the same as differential pulse control, anyway.

  • 1
    $\begingroup$ Edited to address. $\endgroup$ Nov 26, 2018 at 4:02
  • $\begingroup$ Searching Spaceflight 101 article for "2.5" finds what appears to be a relevant section, but I don't really understand it. $\endgroup$
    – uhoh
    Nov 26, 2018 at 4:05
  • 1
    $\begingroup$ That's about the trajectory control thrusters, not the landing thrusters. $\endgroup$ Nov 26, 2018 at 4:06
  • 1
    $\begingroup$ Hm, I think the attitude control and TCM thrusters are attached to the backshell, not the lander, so roll control must be handled by the landing thrusters entirely. $\endgroup$ Nov 26, 2018 at 4:34
  • $\begingroup$ Maybe it has some RCS for the roll though. I don't think they would just launch a probe without controling the roll on descent. Still thank you very much for your answers! $\endgroup$ Nov 27, 2018 at 15:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.